full-scale test piles
TRANSCRIPT
INVESTIGATION OF SOIL DAMPING ON FULL-SCALE TEST PILES
By
Dirk A. van Reenen Research Assistant
Harry M. Coyle Associate Research Engineer
Richard E. Bartoskewitz Engineering Research Associate
Research Report Number 125-6
Bearing Capacity for Axially Loaded Piles Research Study Number 2-5-67-125
Sponsored by The Texas Highway Department
In Cooperation with the U.S. Department of Transportation Federal Highway Administration
August 1971
TEXAS TRANSPORTATION INSTITUTE Texas A&M University
College Station, Texas
The op1n1ons, findings, and con~lusions expressed in this report are those of the authors and not necessarily those of the Federal Highway Administration.
ii
ABSTRACT
Bearing capacity is predicted and compared with field
load test results for six full-scale·test piles. A computer
program for studying piling behavior by wave equation analysis
is used to develop soil damping parameters. Th~se soil damping
parameters are used in a mathematical model which describes
the damping characteristics of the sbil.
For the four test piles which were embedded entirely in
highly plastic clay, a recommended value for the friction
damping parameter and the point damping parameter is established.
The point damping parameter is investigated for two test piles
which were embedded in clay with the tips in sand. The point
damping parameter for piles with the tip in sand is found to
be 1 arger than the recommended va 1 ue for piles embedded entirely
in clay. A reco~~ended value for the point damping parameter
in sands is not established because of the limited amount of
field data available for this study.
iii
SUMMARY
This investigation was conducted during the fourth year
of a five year study on "Bearing Capacity for Axially Loaded
Piles.'' The applicability of two different mathematical models
which describe soil damping characteristics is evaluated for
use in wave equation analyses of piling behavior. A wave
equation computer program is used to predict pile bearing
capacity, and the predicted capacity is compared with field
load test data from full:scale test piles: Wave e~uation
analyses of four full-scale teit piles embedded entirely in
highly plastic clay soils indicate that an average value of
J' = 0.2 seconds per foot for the friction damping and a point
damping parameter of J = 0.15 seconds per foot may be used
\'lith the mathematical model having a velocity exponent of 1.0.
Two additional test sites are analyzed where the test
piles were embedded predominantly in a highly plastic clay,
but the ~ile tips were embedded in sand. Using the mathematical
model with a velocity exponent of 1.0 and a point damping value
of 0.15 seconds per foot, extremely high values of friction
damping are needed to achieve agreement between predicted and
actual bearing capacity. However, if,a friction damping value
of 0.20 seconds per foot is used, point damping values of
0.95 seconds per foot and 1.55 seconds per foot are obtained.
iv
The analysis of these two test piles with tips embedded in
sand indicates that a value of point damping much greater
than 0.15 seconds per foot is required for agreement between
predicted and actual bearing capacity. However, additional
test data are needed to verify this trend.
v
IMPLEMENTATION STATEMENT
This is a technical progress report which presents the
results of an investigation which was conducted to develop
soil damping parameters .. These soil damping parameters are
needed for use with the one-dimensional wave equation computer
program to predict the bearing capacity of an axially loaded pile.
Implementation of the results obtained in this study
should be limited to piles which are embedded entirely in
highly plastic clay soils. A value of J' = 0.2 seconds per
foot for the friction damping parameter and a point damping
parameter of J = 0.15 seconds per foot are recommended for
us~ with the mathematical model having a velocity exponent of
1.0. The values of point damping for piles driven through a
highly plastic clay layer into sand determined in this study
should not be used until further verification has been obtained
·from additional field tests. Future field tests should ·include
the measurement of point load through instrumentation. The
initial static load tests should be conducted at the time of
driving, and the ten-day static load tests should be conducted
concurrently with a redriving of the test piles. This procedure
will allow predicted values of bearing capacity and soil set-up
to be correlated with the actual values obtained in the field.
vi
Implemented results from this study should be used concurrently
with existing design procedures pending further verification I
I
by additional field tests on full-scale piles .
. . I
,!
vii
TABLE OF CONTENTS
INTRODUCTION .. . . . . . . . . Present Status of the Question Objectives .......... .
INVESTIGATION OF FRICTION DAMPING ON PORT ARTHUR TEST PILES .................. .
General . . . . . . . . . . . . . . . . . . . Analysis 6f Friction Damping at Time of Driving Analysis of Friction Damping ll Days after Driving
ADDITIONAL CASE STUDIES OF PILES ALL IN CLAY
General . . . . . . . . . Beaumont Test Pile ..... Chocolate Bayou Test Pile Discussion of Results
CASE' STUDIES OF PILES IN CLAY WITH TIPS IN SAND
General . . . . . . . . . . Victoria Test Pile ..... St. Charles Parish Test Pile Discussion of Results
INVESTIGATION OF POINT DAMPING PARAMETER
General . . . . . . . . . . . Method of Analysis ..... . Discussion of Results .•..
CONCLUSIONS AND RECOMt1ENDATIONS
Conclusions ..... . Recommendations
APPENDIX I.- REFERENCES
APPENDIX II.- SUMMARY OF INPUT DATA
Port Arthur Test Piles ..... .
viii
Page
1
1 4
5
5 5 ~
10
13
13 15
• 15 • 17
. . . 23
. 23 . . 24 . . 25
. 27
. 29
. . 29 • 29 . 30
. 34-
. 34 • 35
. 37
. 39
• 39
TABLE OF CONTENTS (CONTlNUED)
Beaumont Test Pile ..... . Chocolate Bayou Test Pile .. Victoria Test Pile ..... St. Charles Parish Test Pile .
ix
Page
. 41
. 42 • 43 . 44
LIST OF FIGURES
Figure
1 Friction Damping Parameter Versus Dynamic Driving Resistance for Port Arthur Test Pile No. l 7
2 Friction Damping Parameter Versus Velocity Exponent for Port Arthur Test Piles at Time of Driving . . . . . . . . . . . . . . . . . . 9
3 Friction Damping Parameter Versus Velocity Exponent for Port Arthur Test Piles After 11 Days 11
4 Friction Damping Parameter Versus Velocity Exponent for Beaumont Test Pile . . . . . . . . . . 16
5 Friction Damping Parameter Versus Velocity Exponent for Chocolate Bayou Test Pile . . . . 18
6 · Friction Damping Parameter Versus Velocity Exponent for Piles All in Clay ..... 19
7 Soil Resistance Versus Dynamic Driving Resistance for Port Arthur Test Pile No. 1 . . . . . . . 21
8 Friction Damping Parameter Versus Velocity Exponent for Victoria and St. Charles Parish Test Piles . . . . . . . . . . . . . . . . . 26
9 Point Damping Parameter Versus Dynamic Driving Resistance for St. Charles Parish Test Pile . . 31
10 Point Damping Parameter Versus Dynamic Driving Resistance for Victoria Test Pile . . . . . . . 32
X
________________ ..........
INTRODUCTION
Present Status of the Question. - Dynamic pile driving
formulas have been in use for many years to predict the lo.ad
carrying capacity of a pile. Unfortunately, these formulas are I
dependent upon simplifying assumptions which greatly reduce
their accuracy and restrict their application. Isaacs (4)* is
credited with first proposing the occurrence of longitudinal
wave transmission in a pile during driving. Because the solution
is extremely complex, little effort was made to expand this
theory until 1960 when Smith (10) presented a numerical solution
to the wave equation and a mathematical model to simulate pile
soil interaction. Smith described the total soil resistance
mobilized during dynamic loading in the following manner:
R =R [l+(JorJ•)v] udynami c ustati c
where Ru =dynamic; or static soil resistance, pounds;
J =a damping constant for the soil at·the point of
the pil~, seconds per foot;
J• =a damping constant for the soil along the side
of the pile, seconds per foot; I
V = tile instantaneous velocity of a point on the
(1)
*Numbers in parentheses refer to the references 1 isted .in Appendix I.
(The citations on the following pages follow the style of the Journal of Soil Mechanics and Foundation Division, ASCE.)
1
pile at a given time, feet per second.
During the past six years, considerable research has been
directed toward determining representative damping parameters
for various types of soils. Initial laboratory studies were
conducted by Reeves, Coyle,and Hirsch (9). These studies
involved the use of a dynamic loading apparatus to determine
damping characteristics of saturated sands subjected to impact
loading. Coyle and Gibson (3) extended the laboratory investi-
gation and correlated damping parameters with common soil
properties such as void ratio for sands and liquidity index for
clays. These investigations showed that the damping parameter
as determined in the laboratory was not constant for the range
of velocities considered. However, a constant laboratory damping
para~1eter was obtained by modifying the Smlth equation as follows:
R = R [1 + (J or J') vN] O<N.<l.O (2) udynamic ustatic
where N =a power to'vvhich the velocity, V, must be raised for
J or J' to be constant.
Tests were later performed by Korb and.Coyle (5) and Raba
and Coyle (8) on model piles in clay. Using the modified Smith
equation, Korb and Coyle investigated both tip damping, J, and
friction damping, J'. For piles in clay, a value of N = 0.35
was recommended for use with a constant friction damping parameter
of J' = 1.25 seconds per foot,and a value of N = 1.0 was
recommended for use with a constant point damping parameter of
J = 0.15 seconds per foot.
2
- --------------------------------
In 1970, Bartoskewltz. and Coyle (2) obtained static and
dynamic field test data on two instrumented piles in clay. Using
the damping parameters recommended by Korb and Coyle for wave
equation analyses, Bartoskewitz and Coyle compared predicted
bearing capacity vd th measured bearing capacity for these two
instrumented piles. The predicted bearing capacity was
approximately 30 percent less than the bearing capacity measured
during the pile load tests. This rather large discrepancy
between measured and predicted pile capacity suggested that a
value be determined for friction damping which would reduce the
error. Using an N value of 0.35 in the wave equation analyses,
values of friction damping required to give agreement between
predicted and actual pile bearing capacity were J' = 0.535
seconds per foot for test pile No. 1 and J' = 0.67 seconds per
foot for test pile No. 2.
These results indicated that the friction damping parameter
corresponding to an N value of 0.35 was not a constant for the
·clay soils at this test site. Bartoskewitz and Coyle attempted
to relate frictioh damping vdth some soil classification
property; but their findings were inconclusive. It was also
observed that an jnfinite number of combinations of J' and N
could be used to achieve agreement between predicted and actual
pile bearing capacity. At this point in the research program,
it became apparent that information was needed concerning the
relationship between the friction damping parameter, J', and the
3
I
L
velocity exponent, N.
Objectives. -The objectives of this investigation are:
a. To obtain static and dynamic field test data on
full-scale instrumented piles from the literature. . '
b. To determine, by using the field test data and the
one-dimensional wave equation analysis, the soil
damping parameters required to achieve agreement
between predicted and actual pile bearing capacity.
c. To evaluate the applicability of two different
mathematical models used to describe the damping
characteristics of the soil with the one-dimensional
wave equation analysis.
4
INVESTIGATION OF FRICTION DAMPING ON
PORT ARTHUR TEST PILES
General. -The computer program developed by Lowery, Hirsch,
and Samson (7) for solving the one-dimensional wave equation
using Smith's (10) numerical method is employed in this investi
gation.· Any future mention of a wave equation analysis or
solution refers to an analysis or solution obtained by using
this computer program.
Analysis of Friction Dameing at Time of Driving. - In
November, 1969, two instrumented piles were driven and 1 oad
tested near Port Arthur, Texas. Both piles were 16-in. OD,
3/8-in. wall thickness, steel pipes and wer~ driven by a
Link Belt 520 diesel hammer. Test pile No. 1 had a total length
of 67 ft and was driven to. an embedded depth of 64 ft. Test
pile No. 2 had a total length of 78 ft and was driven to an
embedded .depth of 74 ft. Both piles were statically lqad tested 'I
at approximately two hours after driving and again 11 days
after driving. Strain gages at the head and tip of each pile
made it possible to meas~re both tip load (RUP) and total soil
resistance (RUT) for each pile. Complete data on the static
load tests have been reported by Bartoskewitz and Coyle (2).
With the exception of the friction damping parameter, J',
and the velocity exponent, N, the soil parameter values used in
5
this investigation are those recommended by Korb and Coyle (5)
and by Bartoskewitz and Coyle (2). These values are:
Qsi de = 0.10 inch
Qpoint = 0.10 inch
Jpoint = 0.15 seconds per foot
Npoint = 1·0
where Q =the maximum elastic soil deformation, or quake.
Values of RUP and RUT used were those measured duri~g the
static load test.
By using the wave equation computer program and by setting
RUT equal to the static load capacity of the pile at time of
driving, it was possible to determine the friction damping
parameter, J', corresponding to a particular value of the
velocity .exponent, N. For example, the static soil resistance
of Port Arthur test pile No. 1 was 46.2 tons two hours after
driving. The point resistance measured from strain gages was
nine tons. These values were used for RUT and RUP respectivelY.
In order to develop the curve shown in Fig. 1, s·everal
values of J' were selected fora value of N = 0.2, and the
corresponding driving resistances were computed using the wave
equation program. The wave equation computer program calculates
the permanent set of the pile caused by one blow of the hammer.
The reciprocal of the permanent set yields the driving resistance
in blows per unit of net pile movement. Since the actual blow
count for the last increment of driving was known, the J 1
6
t;:: 0:: w 0....
u w (./)
n
.......... -'-::> -0:: w 1-w :2::
-.....! c.:( 0:: o:::( 0....
(.!:5
z:: ,_. 0.... :2:: o:::( 0
z:: 0 ....... 1-u ....... 0:: LJ_
1.6
1.2
0 5
FOR N = 0.2
Measured dynamic driving resistance = 14.5 blows per foot
10 15 20 25 30
DYNAMIC DRIVING RESISTANCE, BLOWS PER FOOT
FIG. 1. -FRICTION DAMPING PARAMETER VERSUS DYNAMIC DRIVING RESISTANCE
FOR PORT ARTHUR TEST PILE NO. 1
value corresponding to N = 0.2 and RUT = 46.2 tons can be
determined. For the recorded driving resistance of 14.5 blows
per foot, the corresponding J' was determined to be 0.70 seconds
per foot. This prdcedure was repeated for values of N = 0.4,
0.6, 0.8, ~nd 1.0. For each value of N, there is a unique value
of J' which forces agreement between the predicted static soil
resistance and the static soil resistance measured in the field.
It was then possible to plot each unique value of J' with the
corresponding N value for Port Arthur test pile No. 1 as shown
in Fig. 2.
Port Arthur test pile No. 2 was. analyzed in the same manner.
The average blow count for the last several feet of driving was
recorded as 16 b 1 ows per foot, and RUT was measured to be 50. 1
tons. The procedure used to develop a relationship between the
side friction damping parameter and the velocity exponent was
identical to that used for pile No. 1. The curve relating J'
to N for Port Arthur test pile No. 2 is also presented in Fig. 2.
Fig .. 2 illustrates the relationship between J' and N for
the range of N values between 0 and 1 .0. For the Port Arthur
test piles, J' approaches a constant value of approximately 0.17
seconds per foot as N increases to 1.0. However, as the value
of N is decreased, the two curves tend to diverge. For example,
according to Bartoskev.Jitz and Coyle (2), if N = 0.35, J' = 0.535
seconds per foot for test pile No. 1 and J' = 0.67 seconds per
foot for test pile No. 2. Thus, it appears that at the time of
8
1-1..1..
0:: LLJ CL
u LLJ (/)
ft
......... -'J ........
0:: LLJ 1-LLJ :::E
\0 ~ c::( CL
(.!j z ....... CL :::E c::( Cl
z 0 ....... 1-u ....... 0:: 1..1..
0.8~------------~--------~--------------~
0.6
0.4
I 0.2
0
PILE NO. 1
0.2 0.4 0.6 0.8 1.0
VELOCITY EXPONENT, N
FIG. 2. - FRICTION DAMPING PARAMETER VERSUS VELOCITY EXPONENT
FOR PORT ARTHUR TEST PILES AT TIME OF DRIVING
driving, the friction damping parameter is fairly constant for a
value of N = l .0 but is not constant for values of N less than
1.0.
Analysis of Friction Damping ll Days after Driving. - The
Port Arthur piles were statically load tested and redriven a
second time 11 days after installation. The difference in the
hammer-pile-soil system between the time the piles were first
driven and the time the piles were redriven after 11 days was
due to set-up occurring· in the clay. This set-up increased
the static soil resistance and decreased the RUP/RUT ratios.
For example, RUP/RUT for Port Arthur test pile No. 1 at time of
driving was 0. 195, but when the pile was redriven after 11 days,
RUP/RUT was 0.050. By using the same procedure described
previously, curves relating the-friction damping parameter
versus driving resistance were obtained for both piles. The
average blow count for pile No. 1 was 72 blows per foot; and
for pile No. 2, 182 b 1 ows per foot. Knowing these b 1 ow counts,
it was possible to develop curves-relating J' and N as shm'ln
in Fig. 3. Comparing Figs. 2 and 3, it is observed that for
any value of N, the corresponding J' is higher at-the -time of
the 11-day test than at the time of driving. For example, Port
Arthur test pile No. 1 at time of driving has a J' value of
0.17 seconds per foot corresponding toN= 1.0. After 11 days,
the J' value corresponding toN= 1.0 is 0.44 seconds per foot.
This increase in J' is due to the set-up which occurs in clays.
10
t: 0:: LLJ c... u LLJ (/) . --'-:> ........ 0:: LLJ ..__ LLJ ::E <( __, 0:: __, <( c... (.!) z ...... c... ::E .::( Cl
z 0 ...... ..__ u ...... 0:: 1.1..
----~--~--~--------~--, 2.0
1.5
1.0
0.5
0
~~ .r-- PILE NO. 2
PILE NO. 1
0.2 0.4 0.6 0.8 l.O VELOCITY EXPONENT, N
FIG. 3. - FRICTION DAMPING PARAMETER VERSUS VELOCITY EXPONENT
FOR PORT ARTHUR TEST PILES AFTER 11 DAYS
Therefore, it is apparent that values for J' obtained at the time
~f drivi~g will not apply after set-up has occurred;
12
ADDITIONAL CASE STUDIES OF PILES ALL IN CLAY
General. -Results from the Port Arthur test piles indicated
that for piles all in clay, J• is approximately 0.17 seconds per
foot when N = 1.0. However, additional data were needed on full
scale pile load tests to determine if this value of the friction
damping parameter would be applicable in the same type of soil
at other locations. Records of test piles driven at Beaumont,
Texas and at Chocolate Bayou, along the Texas Gulf Coast were
ana-lyzed. Both of these piles were statically load tested at
least ten days after driving. Thus, it was necessary to estimate
the amount of set-up which had occurred so that the static soil
resistance at time of driving could be determined.
Only by performing a static load test immediately after a
pile is driven and then again after soil set-up has occurred can
the amount of set-up be determined exactly. However, this
method is impractical because of the time and expense incurred
in conducting two separate tests.
Bartoskewitz and Coyle (2) report a soil set-up of 2.16 and I
2.43 for the Port'Arthur piles; i.e., the static load capacity
of pile No. 1 at the end of 11 days was 2.16 times the static
load capacity of the pile at the time of driving; and for pile
No. 2, the static load capacity increased by a factor of 2.43
between time of driving and 11 days after driving. Tomlinson(ll)
13
has presented data in the form of bearing capacity versus time
curves from which a set-up factor of approximately 2.0 was
suggested. Therefore, in this investigation a set-up factor
of 2.0 was used in the absence of conclusive static load test
data.
In reviewing the original Port Arthur data reported by
Bartoskewitz and Coyle (2), it was concluded that no change in
point resistance with time should be expected. For Port Arthur
pile No. 1, the point resistance decreased from nine tons at
time of driving to five tons at 11 days, and the point resistance
at pile No. 2 increased from eight tons at time of driving to
ten tons at 11 days. Thus, there appeared to be no trend in the
change of point resistance with time, and it was concluded that
·in clay soils point resistan<;:e should be considered to remain
constant with time.
For the cases in which point resistance was not known , it
was assumed to remain constant, and a set-up factor of 2.0 was
applied to side friction. For example, at the time the static
load test was performed, the Chocolate Bayou test pile had a
total static soil resistance of 120 tons and a point resistance
of 20 tons. Assuming that the point resistance remained constant
and that the side friction increased by a factor of 2.0, the
static soil resistance at the time of driving was 70 tons (20
tons plus half of 100 tons). Therefore, RUP/RUT at time of
driving was 20/70 or 28.6%.
14
--------------------------------~~~----------------
Beaumont Test Pile. - Data obtained from a pile load test
conducted at Beaumont, Texas have been reported by Airhart,
·Hirsch, arid Coyle (1). This pile was a 16-in. 00, 3/8-in. wall,
53-ft long steel pipe driven into predominantly clay soils by
a Delmag D-12 hammer. The average blow count for the last
s..everal feet of driving was 28 blows per foot, and the pile was
tested 13 days after driving. The maximum static test load
applied on the pile was 120 tons, and the tip resistance was
measured to be 18 tons. Keeping the point resistance constant
and reducing the side friction by a factor of 2.0, RUT at time
of driving was calculated to be 18 tons plus 102 tons/2.0 = 69
tons. RUP/RUT = 18/69 or 26.1%. Knowing these values of RUP
and RUT and using the soil parameters determined previously, a
wave equation analysis was performecj in _which the friction
damping parameter, J' , was varied between o, and 1. 6 seconds per
foot;and the velocity exponent, N, was varied between 0 and
l .0. The curve of J' versus N is shown in Fig. 4, and it is
observed that for N = l.a, J' = 0.22 seconds per foot.
Chocolate Bayou Test Pile. -Lowery, Edwards, and Hirsch (6)
reported results of a static load test conducted at Chocolate
Bayou on the Texas Gulf Coast. This instrumented pile was a
16-in. square precast concrete pile 40-ft long. The pile was
driven into predominantly clay soils to an embedded depth of 33
ft by a Link Belt 520 hammer. The average blow count the last
several feet of driving was 24 blows per foot, and the pile was
15
I-LL.
0:: LLJ 0...
u LLJ V)
.. ......... -"J .._..
0::: LLJ 1-LLJ :::E c1: 0::: c1: _, 0...
0"1 (.!)
z ,_. 0..
~ 0
z 0 ,_. I-u ,_. 0:: LL.
1.6
1.2
0.8
0.4
0 0.2 0.4 0.6 0.8 1.0
VELOCITY EXPONENT, N
FIG. 4. - FRICTION DAMPING PARAMETER VERSUS VELOCITY
EXPONENT FOR BEAUMONT TEST PILE
load tested 45 days after driving. The maximum static load
applied to the pile was 120 tons; the point resistance was meas
ured to be 20 tons. Since the pile was not tested immediately
after driving, it was again necessary to assume a set-up of
2.0 for side resistance, while the point resistance was assumed
to remain constant. RUT was found to be 20 tons plus 100 tons/
2.0 = 70 ton~ RUP/RUT = 20/70 or 28.6%. Following the same
procedure described for the Beaumont pile, the curve of J'
versus N shown in Fig. 5 was obtained. For N = 1.0, the
corresponding J' was found to be 0. 26 seconds per foot.
Discussion of Results. -The rel~tionships between J' and
N for the four test piles investigated are summarized in Fig. 6.
These relationships were developed for values of N between 0
and 1.0. It is observed from Fig.~ that for anN value of 1.0, '.
the friction damping parameter converges to an average value of
b.20 seconds per foot. In each case, the piles were driven into
predominantly highly plastic clay soils. Therefore, the
·following soil parameter values are recommended for piles driven
entirely into highly plastic clays:
Friction damping, J' = 0.20 seconds per foot,
Point damping, J = 0.15 seconds per foot,
Velocity exponent, N = 1.0.
By recommending a value of N = 1.0, it is implied that
Smith's equation, Eq. l, need not be modified and should remain:
Rudynamic = RUstatic [l + (J or J') V].
17
....,... LL.
0:::: w o._
u w (/')
" ........... -'J --0:::: w 1-w :a: ~ c::C CL -co ~ z ,_. o._ :a: c::C a z 0 ...... 1-u ...... 0:::: LL.
1.6
1.2
0.8
0.4
0 0.2 0.4 0.6 0.8 1.0
VELOCITY EXPONENT, N
FIG. 5. - FRICTION DAMPING PARAMETER VERSUS VELOCITY EXPONENT
FOR CHOCOLATE BAYOU TEST PILE
..
1.6
1-LL.
0::: L1J 0...
u 1.2 L1J (/')
" ..-.... -"":) .__
0::: L1J 0.8 1-L1J ::E c:( 0::: c:(
1.0 0...
<.!l z ...... 0... 0.4 ::E c:( Cl
z 0 ...... 1-u ...... 0::: LL.
0
\ 1 -- Beaumont Test Pile 2 -- Chocolate Bayou Test Pile 3 -- Port Arthur Test Pile No. 2 4 -- Port Arthur Test Pile No. 1
0.2 0.4 0.6 0.8 1.0
VELOCITY EXPONENT, N
FIG. 6. ~ FRICTION DAMPING PARAMETER VERSUS VELOCITY EXPONENT
FOR PILES ALL IN CLAY
The soil parameter values listed above were used with the
wave equation computer program to determine predicted bearing
capacity for the four piles previously discussed. In order to
estimate the static soil resistance; the following procedure
was employed. By using the above soil parameters and by varying
RUT, a curve of RUT versus dynamic driving resistance was
obtained. Fig. 7 shows the relationship bet\veen static soil
resistance and dynamic d_riving resistance for Port Arthur test
pile No. 1. Entering the graph with the known blow count of
14.5 blows per foot, the corresponding predicted RUTwas deter,;..
. mined to be 41.0 tons. The RUT actually measured during the
static load test was 46.2 tons; thus, the error between the
predicted and actual bearing capacity is -5.2 tons/46.2 tons or I
-11.2%. This procedure was repeated for each of the four test
piles investigated, and the result~ ar~ summarized in Table 1.
The
Table 1. -Error in Predicting Static Soil Resistance
Caused by Using an Average Value of J'
Test Pile Measured Soil Predicted Soi 1 Error Location Resistance Resistance (%)
(tons) (tons)
Port Arthur No. 1 46.2 41.0 -11.2
Port Arthur No. 2 50.1 46.4 - 7.4
Beaumont 69.0 73.0 + 5.8
Chocolate Bayou 70.0 76.0 + 8.6
maximum difference between predicted and actua 1 bearing
20
'..II z 0 I-. d z < I-V) _, './)
N LJ.J ...... 0:::
_J ...... 0 './)
u -!;;( I-'./)
80,_------------~------------------------------------------~
60
40
20
0 5 10 15 20 25
DLnMIC DRIVING RESISTANCE, BLOWS PER FOOT
FIG. 7.- SOIL RESISTANCE VERSUS DYNAMIC DRIVING RESISTANCE FOR
PORT ARTHUR TEST PILE NO. 1
30
capacity is in the order of plus or minusten percent which is
considered acceptable agreement.
22
CASE STUDIES OF PILES IN CLAY WITH
TIPS IN SAND
General. - Additional test pile records were located in
which the piles were driven through an extensive upper clay
layer and into several feet of sand. These test pile sites
were located at St. Charles Parish, Louisiana and at Victoria,
Texas. The clay soils which provided side resistance during
load testing were very similar to the clay soils found at the
sites presented previously. However, the tip resistances of
the piles driven several feet into ~and were much higher than
the tip resistances of piles driven entirely into clay.
As was the case in the Beaumont and Chocolate Bayou test
piles, the load tests at Victoria and St. Charles Parish were
conducted a minimum of ten days after the pile was driven.
Thus, the determination of soil resistance at time of driving
was required.
On May 10, 1971, the Texas Highway Department drove
a 16-in. square prestressed concrete pile on Park Road 22
near Corpus Christi, Texas. The pile was instrumented by
TTl for research purposes. The soil profile indicated that
the test pile was embedded almost entirely in sand. Load
tests were conducted immediately after driving and again
23
ten days after driving. Strain gages at the head and tip of the
pile made possible the measurement of total static soil resist
ance and tip resistance. At time of driving, thetotal static
soil resistance was 147 tons, and the tip resistance was 109
.tons. Ten days later the st~tic load test was repeated~ and
the total resistance was 156 tons \vith a corresponding tip load
of 108 tons. These uripubl i shed data indicate no change in tip
resistance With time for pile tips embedded in sand. ·Therefore,
it is believed that ultimate point resistance should be con
sidered to remain constant with time for piles embedded in sand.
In this study, point resistance was assumed to remain constant,
and a set-up of 2.0wasapplied to side friction in the clay
layer.
Victoria Test Pile. - Static load tests on an instrumented
test pile at Victoria,_ Texas have been reported by Lowery,
Edwards, and Hirsch (6). The 45-ft long 16-in. square precast
concrete pile was driven to an embedded depth of 30 ft by a
Vulcan No. l hammer. The pile was driven through predominantly
clay soils, and the tip was embedded 4ft into sand. For the
last foot of driving, the blow count was observed to be 395
blows per foot. The static load test was conducted 45 days
after driving. Results from the static load test indicated a
total static soil resistance of 200 tons, and measurements from
the strain gage at the tip of the pile indicate~ a tip load of
128 tons. Therefore, the side resistance was 200 tons minus
24
128 tons or 72 tons. Assuming the point resistance remained
constant with time and applying a set-up of 2.0 to side friction,
the static resi~tance at time of driving was calculated to be
128 tons plus 72 tons/2.0 = 164 tons. RUP/RUT = 128/164 or
78.0%.
Following the same procedure used in analyzing the test
piles embedded entirely in clay, relationships were developed
between the friction damping parameter, J', and the velocity
exponent, N, for a range of N values from 0 to 1.0. The curve
of J' versus N shown in Fig. 8 indicates that for a value of
N = 1.0, the corresponding value of J' is 2.75 seconds per foot.
This value of J' is extremely high compared with the average J'
of 0.20 seconds per foot determined from test piles embedded
entirely in clay. There is no apparent reason for the high
friction damping parameter as the claY soils at the Victoria
site were very similar to those encountered in Port Arthur,
Beaumont, and Chocolate Bayou.
St. Charles .Parish Test Pile. - A w.ave equation analysis was
a 1 so performed on unpub 1 i shed 1 oad test data obtai ned from the
Louisiana Department of Highways. The test pile was a 91-ft
long 54-in. diame~er precast concrete pile with a 5-in. wall
thickness. A Raymond 8/0 hammer was used to drive the pile to
its embedd~d depth of 89 ft. The soil profile was similar to
that encountered at the Victoria site; the pile was driven
through the upper clay layer and embedded 4 ft in sand.
25
1-u. ~ I.J..J Cl..
u I.J..J V)
A.
........... -'-:> ~
~ I.J..J 1-I.J..J ::2: o:::( ~
N o:::( 0'1 Cl..
<-'J z I-t
Cl.. ::2: o:::( Cl
z 0 I-t
1-u I-t
~ LL
8.0 __ ~----------'----------,
6.0
I
4.0
I
2.0
0
~~ VICTORIA TEST PILE
ST. CHARLES PARISH TEST PILE
0.2 0.4 0.6 0.8 1.0
VELOCITY EXPONENT, N
FIG. 8. --FRICTION DAMPING PARAMETER VERSUS VELOCITY EXPONENT FOR
VICTORIA AND ST. CHARLES PARISH PILES
-------------------~~~-~-----
Because the pile was not instrumented, a static analysis was
carried out to determine the distribution of soil resistance
on the pile. The total static soil resistance was measured to
be 625 tons, and the side resistance obtained from a static
analysis was found to be 233 tons. Thus, the point resistance
was 625 tons minus 233 tons or 392 tons. Assuming the point
resistance remained constant with time and applying a set-up
of 2.0 to the side resistance, RUT was calculated to be 392 tons
plus 233 tons/2.0 or 509 tons. The ratio RUP/RUT = 392/509 or
77.2%. The wave equation computer program was used to determine
the relationship between friction damping and the velocity
exponent, and the curve of J• versus N for the St. Charles
Parish test pile is plotted in Fig. 8. It is observed that
for a value of N = 1.0, the corresponding value of J• is 2.74
seconds per foot. Again, this value is considered to be very
high for the clay soils through which most of the pile was driven.
Discussion of Results. -Although the Victoria and St.
Charles -Parish test piles were driven into predominantly clay
soils, the last 4 ft of the piles were embedded in sand.
Values of friction damping corresponding to anN value of 1.0
were 2.75 seconds per foot and 2.74 seconds per foot respectively.
These high values were not expected since it seems logical that
the friction damping parameter should remain constant in the
same type of soil.
The significant difference between the Victoria and
27
St. Charles Parish test piles and the piles embedded entirely
in clay is the high RUP/RUT ratios resulting from the pile tips
being driven into sand. It is possible that the value of the
fri~tion damping parameter for the Victoria and St. Charles
Parish test piles is approximately 0.20 seconds per foot as
indicated from the analysis of piles embedded all in clay.
Furthermore, it is possible that by driving the pile tips into
sand and creating high RUP/RUT ratios, the point damping
parameter is no l anger equal to 0. 15 seconds per foot. · For
this reason, further investigation was directed toward the
relationship between the point damping parameter, J, and the
ratio RUP/RUT.
28
INVESTIGATION OF POINT DAMPING PARAMETER
General. - An investigation of the point damping parameter,
J, was also conducted using the wave equation program. Results
from the friction damping investigation presented earlier indi
cate that a value of N = 1.0 and a corresponding value of J' =
0.20 seconds per foot should be used for piles embedded entirely
in highly plastic clays. Since the Victoria and St. Charles
Parish soils fall in this category, those values were used in
the study of point damping. With the exception of the point
damping parameter, all other parameters were the same as those·
used in the fri,ction damping investigation. The point damping
parameter, J, was varied from 0 to 2.0 seconds per foot, and
the corresponding N value remained a constant value of l.d.
Method of Analysis. -By setting RUT equal to the static
load capacity of the pile at time of driving and by using the·
soil parameters described above., it was possible to determine
the value of J corresponding to a particular ratio of RUP/RUT.
For example, the total static soil resistance for the St. Charles
Parish test pile at time of driving was calculated to be 509 tons
with a point resistance of 392 tons. Using these values for RUT
and RUP respectively and using an initial value of J = 0.5
seconds per foot, the wave equation program was used to compute
the driving resistance in blows per unit of net pile movement.
29
This procedure was repeated for values of J = 1.0, 1.5, and 2.0
seconds per foot. A graph of J versus dynamic driving resistance
for the St. Charles Parish test pile is shown in Fig. 9. Since
the actual blow count for the last increment of driving was
known, Fig. 9 was used to determine the required J value
corresponding to an RUT of 509 tons. For a recorded driving
resistance of 235 blows per foot, the corresponding point
damping parameter wa~ found to be 1.55 seconds per foot. Thus,
the value of the point damping parameter corresponding to the
RUP/RUT value of 77.2% was 1.55 seconds per foot.
This procedure was repeated for the Victoria test pile,
with RUP/RUT = 78.0%, and a J value of 0.95 seconds per foot
was obtained. The graph of J versus dynamic driving resistance
for the Victoria test pile is shown in Fig. ·10.
Discussion of Results. - From the friction damping
investigation, it seems possible that a relationship exists
between the point damping parameter and RUP/RUT. However, results
from the St. Charles Parish and Victoria test piles .do little to
substantiate this premise. Values of friction damping were
determined which give agreement between predicted and actual pile·
bearing capacity. For approximately the same value of RUP/RUT,
0.78, the corresponding values of J were far apart, 1.55 seconds
per foot for the St. Charles Parish test pile and 0.95 seconds
per foot for the Victoria test pile. It should be noted, however,
that both J values were much high·er than the J value of 0.15
30
f-I.J....
~ w 0....
u w (/')
'J ...._,
~ w f-w ::E: < w ~ I-' < 0....
(.!J ·Z: ...... 0.... ::;;:: < Cl
f-z: ...... 0 0....
2.0·r-----------------~--~----------------------
1.5
1.0
0.5
0
----------------------------
50 100 150 200 250
DYNAMIC DRIVING RESISTANCE, BLOWS PER FOOT
FIG. 9. - POINT DAMPING PARAMETER VERSUS DYNAMIC DRIVING
RESISTANCE FOR ST. CHARLES PARISH TEST PILE
300 350
c
I-LJ_
0:: L1..J 0..
u L1..J (/) . ........ '-:> ..__.,
0::: L1..J I-w :a: ~ 0::: ~ 0..
w (..')
N z: ...... 0.. :a: ~ 0
I-z: ...... 0 0..
1.6
1.2
0.8
0.4
0
------------------------I I
"I I I I I I I I I I I
100 200 300 400 500
DYNAMIC DRIVING RESISTANCE, BLOWS PER FOOT
FIG. 10. - POINT DAMPING PARAMETER VERSUS DYNAMIC DRIVING
RESISTANCE FOR VICTORIA TEST PILE
600 700
"'
seconds per foot used in the friction damping investigation.
Too few test piles were analyzed to completely discount the
possibility that a relationship between point damping and RUP/RUT
exists. Additional data for piles driven primarily into clay
and embedded in sand are needed before definite conclusions can
be drawn as to what effect, if any, RUP/RUT has on the point
damping parameter.
33
CONCLUSIONS AND RECOMMENDATIONS
Conclusio~~· - The primary objectives of this investigation
were twofold. First, using the wave equation computer program,
an investigation was made to determine the soil damping parameter
required to achieve agreement between predicted and actual pile
bearing capacity. Second, the results achieved in the first
objective were used to determine which of two mathematical models
better describes the friction damping characteristics of the
soi 1.
Based on wave equation analyses of the Port Arthur,
Beaumont, and Chocolate Bayou test piles at time of driving,
the following conclusions can be made for piles embedded
entirely in highly plastic clays:
1. Smith's mathematical model better describes the
total soil resistance during dynamic loading; i.e.,
the velocity, V, should be raised to a power of
N = l.d.
2. A value of 0.20 seconds per foot should be used
for the friction damping parameter and a value
of 0.15 seconds per foot for the point damping
parameter.
3. Differences in pile materials and oile geometry
do not seem to affect the friction damping parameter.
34
Based on wave equation analyses of the St. Charles Parish
and Victoria test piles at time of driving, the followinq
observations are made for piles driven through a layer of highly
plastic clay with tips founded in sand:
1. For the case in which a point damping parameter,
J, of 0.15 seconds per foot was used, unr~asonable
values of the friction dampinq parameter, a•, were
obtained; i.e., J' = 2.74-2.75 seconds per foot.
2. Using a value of J' = 0.20 seconds per foot as
determined in the friction damping investigation,
higher values of point dampina were obtained;·
i.e., J = 0.95 seconds per foot and 1.55 seconds
per foot.
The attempt to determine a relationship between point
damping and RUP/RUT yielded inconclusive results. However,
for the limited number of cases analyzed, there are indications
that a value of point damping much greater than 0.15 seconds per
foot is required for piles with·tips founded in sand. ·
Recommendati~~· - The various pile analyses presented in
this study are based on piles driven primarily into clay soils.
There is great need for additional field test data obtained from
instrumented piles driven entirely into cohesionless materials
and instrumented piles driven through an upper clay layer with
their tips embedded in cohesionless soils.
It is recommended that future pile tests include a
35
measurement of the tip resistance whenever possible.
Where practical, future static load tests should also be
conducted as soon after driving as possible to evalute soil
pa~ameters at time of driving. A second static load test should
be conducted concurrently with redriving the pile a minimum of
ten days after initial driving. Data from the second static
load test will be useful in evaluating soil parameters after
set~up has occurred.
A great deal more field test data on instrumented piles is
needed in order to thoroughly evaluate the friction and point
damping parameters in all soils and combinations of soils.
36
APPENDIX I. -REFERENCES
l. Airhart, T.P., Hirsch, T.J., and Coyle, H.M._, 11 Pile-Soil System Response in Clay as a FunCtion of Excess Pore Water Pressure and Other Soil Properties," Texas Transportation Institute, Piling Behavior Research, Research Report 33-8, Texas A&M University, September, 1967. _ ·
2. Bartoskewitz, R.E., and Coyle, H.M., ''~vave Equation Prediction of Pile Bearing Capacity Compared with Field Test Results/ Texas Transportation Institute, Bearing Capacity for Axially Loaded Piles, Research Report No. 125-5, Texas A&M University, December, 1970.
3. Coyle, H.M., and Gibson, G.C., "Empirical Damping Constants for Sands and Clays," Journa 1 of the Soi 1 Mechanics and Foundations Division, ASCE, Vol. 96, No. SM3, Proc. Paper 7296, May, 1970, pp. 949-965.
4. Isaacs, D.V., "Reinforced Concrete Pile Formula," Transactions of the Institution of Engineers, Australia, Vol. 12, 1931, pp. 312-323.
5. Korb, K.fL, and Coyle, H.M., 11 Dynamic and Static Field Tests on a Small Instrumented Pile," Texas Transportation Institute, Bearing Capacity for Axially Loaded Pi 1 es, Research Report No. 125-2, Texas A&M University, February, 1969.
6. Lowery, L.L., Edwards, T.C., and Hirsch, T.J., "Use of the Wave Equation to Predict Soil Resistance on a Pile During Driving," Texas Transportation Institute Research Report 33-10, Texas A&M University, August, 1968.
7. Lowery, L.L., Hirsch, T.J., and Samson, C.H., "Pile Driving Analysis- Simulation of Hammers, Cushions, Piles, and Soil," Texas Transportation Institute Research Report 33-9, Texas A&M University, August, 1967.
8. Raba, C.F., Jr., and Coyle, H.M., "The Static and Dynamic Response of a Miniature Friction Pile in Remolded Clay," paper presented at Texas Section, ASCE, San Antonio, Texas, October, 1968.
37
9. Reeves, G.N., Coyle, H.M., and Hirsch, T.J., "Investigation of Sands Subjected to Dynamic Loading," Texas Transportation Institute, Piling Behavior Research, Research Report No. 33-7A, Texas A&M University, December, 1967.
lq. Smith, E.A.L., "Pile Driving Analysis by the Wave Equation," Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 86, No. SM4, Proc. ·Paper 2574, August, 1960, pp. 36-61.
11. Tomlinson, M.J., Foundation Design and Construction, John Wiley and Sons, New York, N.Y., 1963, p. 371.
38
0
APPENDIX II.- SUMMARY OF INPUT DATA
Port Arthur Test Piles
Site No. 1
Hammer Properties
Type: Link Belt 520 diesel
Ram ve 1 ocity: · 14. 70 f~s
Ram weight: 5.07 kips
Anvil weight: 1.18 kips
Adapter weight: 1 . 05 kips
Capblock stiffness: 108,500 kips/in.
Cushion stiffness: 18,600 kips/in.
Capb·1ock coefficient of restitution: 1.0
Cushion coefficient of restitution: 0.8
Pile Properties
Type: 16-in. 00, 3/8-in. wall, steel pipe
Pile length: 67ft
Embedded depth : 64 ft
Segment length: 5 ft
Segment weight: 0.313 kip
Segment stiffness: 9,080 kips/in.
Soil Properties
Static soil resistance: 92.4 kips
Point resistance: 18.0 kips
Final blow count: 14.5 blows/ft
39
Site No. 2
Hammer Properties
Type: Link Belt 520 diesel
Ram velocity: 15.16 fps
Ram weight: 5.07 kips
Anvil weight: 1.18 kips
Adapter weight: 1.05 kips
Capb1ock stiffness: 108,500 kips/in.
Cushion stiffness: 18,600 kips/in.
Capblock coefficient of restitution:
Cushion coefficient of restitution:
Pile Properties
1.0
0.8
Type: 16-in. OD, 3/8-in. wall, steel pipe
·Pile length: 78ft
Embedded depth: 74 ft
Segment length: 5 ft
Segment weight: 0.313 kip
Segment stiffness: 9,080 kips/in.
Soil Properties
Static soil resistance: 100.2 kips
Point resistance:
Final blow count:
40
16.0 kips
16 blows/ft
Beaumont Test Pile
Hammer Properties
Type: ·oelmag D-12 diesel
Ram velocity: 21 .10 fps
Ram weight: 2.75 kips·
Anvil weight: 0.816 kip
Adapter weight: 0.597 ki~
Capblock stiffness: 31,500 kips/in.
Cushion stiffness: 18,600 kips/in.
Capblock coefficient ~f restitution: 1.0
Cushion coeffi~ient of restitution: 1.0
Pile Properties
Type: 16-in. 00, 3/8-in. wall, steel pipe
Pile length: · 53 ft
Embedded depth: 50 ft
Segment length: 5 ft
Segment weight: 0.290 kip
Segment stiffness: 8.,780 kips/in.
Soil Properties
Static soil resistance: 138 kips
Point resistance~ 36 kips
Final b)ow count: 28 blows/ft
41
Chocolate Bayou Test Pile
Hammer Properties
Type: Link Belt 520 diesel
Ram ve 1 oci ty: 11 . 63 ~ps
Ram weight: 5.07 kips
Anvil weight~ 1.18 kips
Adapter weight: 1.30 kips
Capblock stiffness: 108,500 kips/in.
Cushion stiffness: 18,600 kips/in.
Capblock coefficient of restitution: 0.8
Cushion coefficient of restitution:
Pile Properties
Type: 16-i n. square precast concrete
Pile length: 40 ft
Embedded depth: 33 ft
Segm~nt length: 5 ft
Segment weight: 1.378 kips
Segment stiffness: 33,100 kips/in.
Soil Properties
Static soil resistance: 140 kips
Point resistance: 40 kips
Final blow count: 24 b1ows/ft
42
0.5
.0.
0
Victoria Test Pile
Hammer Properties
Type: Vulcan No. 1 steam
Ram velocity: · 12.80 fps
Ram weight: 5.00 kips
Adapter weight: l .00 kip
Capblock stiffness: 1 ,492 kips/in.
Cushion stiffness: 1.736 ki~s/in.
Capblock coefficient of restitution: 0.5
Cushion coefficient of restitution: 0.5
Pile Properties
Type: 16-in. square precast concrete
Pile length: 45 ft
Embedded depth: 30 ft
Segment length: 5 ft
Segment weight: 1.20 kips
Segment stiffness: 37~600 kips/in.
·Soil Properties
Static soil resistance: 328 kips
Point resistance: 256 kips
Final blow count: 395 blows/ft
43
St. Charles Parish Test Pile
Hammer Properties
Type: Raymond 8/0 steam
Ram velocity: 12.94 fps
Ram weight: 25.00 kips
Adapter weight: 6.00 kips
Capblock stiffness: Q,600 kips/in.
Cushioh stiffness: 5,180 kips/in.
Capblock coefficient of restitution: 0.8
Cushion coefficient of restitution: 0.5
Pile Properties
Type: 54-in. cyclinder precast concrete, 5-in. wall
Pile length: 91 ft
Embedded depth: 89 ft
Segment length: 5 ft
Segment weight: 4.19 kips
Se~ment stiffness: 77,800 kips/in.
Soil Prope1~ties
Static soil resistance: 1,018 kips
Point resistance: 785 kips
Final blow count: 235 blows/ft
44